Biomedical Engineering Reference
In-Depth Information
7
MEDICAL IMAGE SEGMENTATION BASED
ON DEFORMABLE MODELS
AND ITS APPLICATIONS
Yonggang Wang, Yun Zhu, and Qiang Guo
Institute of Image Processing and Pattern Recognition
Shanghai Jiaotong University, Shanghai, China
Deformable models, including parametric deformable models and geometric deformable
models, have been widely used for segmenting and identifying anatomic structures in medi-
cal image analysis. This chapter discusses medical image segmentation based on deformable
models and its applications. We first study several issues and methods related to medical
image segmentation and then review deformable models in detail. Three applications in
different medical fields are introduced: tongue image segmentation in Chinese medicine,
cerebral cortex segmentation in MR images, and cardiac valve segmentation in echocardio-
graphic sequences.
1.
MEDICAL IMAGE SEGMENTATION
1.1. Background
Image segmentation plays a crucial role in image analysis. Its goal is to
partition an image into non-overlapping and meaningful regions that are uniform
with respect to certain characteristics, such as graylevel, color, and texture.
Medical image segmentation is becoming an increasingly indispensable step
in image processing for identifying tissue organization from the human body in
numerous medical imaging modalities, including X-ray Computed Tomography
(CT), Magnetic Resonance (MR), Positron Emission Tomography (PET), and
